Stanford Medicine researchers developed and validated a machine‑learning model that predicts whether donation‑after‑circulatory‑death (DCD) donors will die within the timeframe needed for viable liver procurement. Reported in The Lancet Digital Health, the model outperformed surgeon judgment and reduced futile procurements—where surgical preparations begin but the donor dies too late—by about 60% in multicenter testing. Authors say the model can prioritize donor‑recipient matching and reduce wasted operating room time and logistical costs, potentially increasing usable DCD organ yields. The study highlights AI’s immediate utility in transplant logistics and organ‑allocation workflows.